swift-testing-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface common to code analysis tools.
- Ingestion points: The skill explicitly processes external
.swiftsource code files provided by the user or found in the project scope, as defined in the 'Hard gates' section ofSKILL.md. - Boundary markers: There are no explicit instructions to use delimiters or 'ignore embedded instructions' blocks to isolate the content of the reviewed files from the agent's operational instructions.
- Capability inventory: Across all provided files (
SKILL.md,references/async-testing.md,references/expect-macro.md,references/organization.md,references/parameterized.md), no scripts or commands containing subprocess calls, dynamic execution (eval/exec), file-write, or network operations were found. The skill relies on the agent's baseline text analysis capabilities. - Sanitization: The instructions do not prescribe any sanitization, filtering, or escaping of the content found within the reviewed Swift files before the agent processes them.
Audit Metadata